LUT-Based Convolutional Tsetlin Machine Accelerator With Dynamic Clause Scaling for Resources-Constrained FPGAs
Open Access
- 23 March 2026
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal on Exploratory Solid-State Computational Devices and Circuits
- Vol. 12, 45-55
Abstract
The rapid growth of machine learning (ML) workloads, particularly in computer vision applications, has significantly increased computational and energy demands in modern electronic systems, motivating the use of hardware accelerators to offload processing from general-purpose processors. Despite advances in computationally efficient ML models, achieving energy-efficient inference on resource-constrained edge devices remains a significant challenge. The Tsetlin Machine (TM) has emerged as an attractive alternative for image classification due to its high throughput and inherently energy-efficient learning paradigm. However, existing TM-based hardware accelerators struggle to balance classification accuracy and energy efficiency, limiting their practical deployment at the edge. This paper presents a resource- and energy-efficient Convolutional Tsetlin Machine (CTM) accelerator with dynamic clause scaling, optimized explicitly for edge Field-Programmable Gate Array (FPGA) platforms. The proposed architecture employs LUT-based pipelining and targeted resource-optimization techniques to minimize FPGA resource utilization while maintaining high energy efficiency and performance. The accelerator is implemented on a Xilinx Zybo-Z20 FPGA and evaluated using the MNIST, Fashion-MNIST (FMNIST), and Kuzushiji-MNIST (KMNIST) datasets, achieving classification accuracies of 97.78%, 85.53%, and 88.54%, respectively, with an energy consumption of up to 0.3 μJ per image classification. Compared with state-of-the-art CTM accelerators, the proposed design achieves up to 40× improvements in resource and energy efficiency, demonstrating its suitability for real-time image and pattern classification on edge FPGA-based systems.Keywords
Funding Information
- Institute of Embedded Systems, University of Siegen, Germany
This publication has 22 references indexed in Scilit:
- Towards Resource Efficient and Low Latency CNN Accelerator for FPGAs: Review and Evaluation2024
- BiDSRS: Resource Efficient Real Time Bidirectional Super Resolution System for FPGAs2024
- FPGA-Based Real-Time Object Detection and Classification System Using YOLO for Edge ComputingIEEE Access, 2024
- Convolutional Tsetlin Machine-based Training and Inference Accelerator for 2-D Pattern ClassificationMicroprocessors and Microsystems, 2023
- BRAMAC: Compute-in-BRAM Architectures for Multiply-Accumulate on FPGAs2023
- Energy-Efficient Precision-Scaled CNN Implementation With Dynamic Partial ReconfigurationIEEE Access, 2022
- A fast and scalable architecture to run convolutional neural networks in low density FPGAsMicroprocessors and Microsystems, 2020
- An End-to-End Solution to Autonomous Driving Based on Xilinx FPGA2019
- Virtualized FPGA Accelerators for Efficient Cloud Computing2015
- Going deeper with convolutions2015